A machine vision technology detection method based on depth learning

Abstract

The invention relates to a machine vision technology detection method based on depth learning, which relates to the technical field of industrial detection. A large number of defective product pictures are collected as samples by an industrial camera under a good industrial light source; These sample pictures are calibrated by software to mark the position of the defects concerned and the types ofdefects. The sample pictures marked with the defect types are used as the material hook part depth learning network model, and the depth learning network model is trained; The trained in-depth learning network model is imported into machine vision system, so as to identify various defects on the spot, and complete the sorting of defective products with industrial automation equipment. To avoid missed detection and false detection, new samples can be learned to improve the accuracy, adaptability and stability, and realize the expert system to meet the needs of various complex scenarios of industrial detection.

Description

technical field [0001] The invention relates to the technical field of industrial detection, in particular to a detection method of machine vision technology based on deep learning. Background technique [0002] The concept of deep learning originated from the study of artificial neural networks. A multi-layer perceptron with multiple hidden layers is a deep learning structure. Deep learning combines low-level features to form more abstract high-level representation attribute categories or features to discover distributed feature representations of data. Deep learning is a method based on representation learning of data in machine learning. It is a new field in machine learning research. Its motivation is to establish and simulate the neural network of human brain for analysis and learning. It imitates the mechanism of human brain to interpret data such as images, sounds and text. Observations (such as an image) can be represented in a variety of ways, such as a vector of...

AI Technical Summary

Problems solved by technology

[0008] 1. Manual extraction of features is required, and the distinction between different categories cannot be fully expressed. For the situation where there are many types of defects and the differences are not obvious, the classification accuracy is low, and it is easy to misdetect and miss;

[0009] 2. It is greatly affected by light, material, background noise and other working conditions, and its adaptability and stability are poor;

[0010] 3. Defects need to be detected first, and then classified, which cannot be done at the same time, and the efficiency is low;

[0011] 4. It is impossible to accurately score and rate problems such as defects and quality

Images